1. Field of the Invention
The present invention relates to a display device and a driving method thereof, particularly to a display device to which a time gray scale method is applied.
2. Description of the Related Art
In recent years, a so-called self-luminous display device in which a pixel is formed using a light emitting element such as a light-emitting diode (LED) has attracted attention. As a light emitting element used for such a self-luminous display device, an organic light emitting diode (OLED) (also referred to as an “organic EL element”, an “electroluminescence (EL) element”, or the like) has attracted attentions, and have been used for an EL display or the like. A light emitting element such as an OLED is of self-luminous type; therefore, it has advantages such as higher visibility of pixels, no backlight, and higher response speed compared to a liquid crystal display. The luminance of a light emitting element is, in addition, controlled by a current value flowing therein.
As a driving method of controlling light emission gray scales of such a display device, there are a digital gray scale method and an analog gray scale method. In the digital gray scale method, a light emitting element is turned on/off by controlling in a digital manner to express gradation. On the other hand, in the analog gray scale method, there are a method of controlling the emission intensity of a light emitting element in an analog manner and a method of controlling the emission time of a light emitting element in an analog manner.
In the case of the digital gray scale method, there are only two states of a light emitting state and a non-light emitting state so that only two gray scale levels can be expressed. Therefore, multi-gray scale display is achieved by combining with another method. As the method for achieving multi-gray scale, a time gray scale method is used in many cases.
As a display in which a display state of a pixel is controlled in a digital manner and a time gray scale method is combined to express gradation, there are some displays other than an organic EL display using a digital gray scale method, such as a plasma display.
A time gray scale method is a method for expressing gradation by controlling the length of a light emitting period and the number of light emissions. That is, one frame is divided into a plurality of subframes, each of which is weighted such as by the number of light emissions or a light emitting period, and the total weight (the sum of the number of light emissions or the sum of the light emitting periods) is differentiated per gray scale level, thereby gradation is expressed. It is known that a display defect called a pseudo contour (or a false contour) occurs when such a time gray scale method is used. Thus, a countermeasure against the problem has been considered (see Patent Document 1).
In addition, the frame frequency has been increased to reduce the pseudo contour. As one of methods, there has been a method in which the length of a subframe is reduced to half so that the number of subframes within one frame is doubled. This is substantially the same as that the frame frequency is doubled (see Patent Document 2). This method is referred to as a “double speed frame method” in this specification.
Here, considered is a case of a 5-bit display (32 gray-scale levels). First, a selection method of subframes according to a conventional time gray scale method, that is, whether each subframe is for lighting or not at each gray-scale level is shown in FIG. 43. In FIG. 43, one frame is divided into 5 subframes (SF1 to SF5) and respective lengths of lighting periods of the subframes are set such that SF1=1, SF2=2, SF3=4, SF4=8, and SF5=16; that is, each length of the lighting period is power of two. Note that a gray scale level of 1 and a length of 1 of a lighting period correspond to each other. By combining these lighting periods, a display with 32 gray-scale levels (a 5-bit gray scale) can be performed.
Here, a way to see FIG. 43 is described. Lighting is performed in a subframe indicated by O-indication whereas lighting is not performed in a subframe indicated by x-indication. Gradation is expressed by selecting a subframe to perform lighting at each gray scale level. For example, in the case of a gray scale level of 0, lighting is not performed in SF1 to SF5. In the case of a gray scale level of 1, lighting is not performed in SF2 to SF5 whereas lighting is performed in SF1. In the case of a gray scale level of 7, lighting is not performed in SF4 and SF5 whereas lighting is performed in SF1 to SF3.
Next, shown in FIG. 44 is an example in which a double speed frame method is applied to the case of FIG. 43. Each subframe in FIG. 43 is divided into two equally, thereby 10 subframes (SF1 to SF10) are formed and respective lengths of lighting periods thereof are such that SF1=0.5, SF2=1, SF3=2, SF4=4, SF5=8, SF6=0.5, SF7=1, SF8=2, SF9=4, and SF10=8. As a result of this, the frame frequency is doubled substantially.
Further, a case of a 6-bit display (64 gray-scale levels) can also be considered similarly. Shown in FIG. 46 is an example in which a double speed frame method is applied to a subframe structure for a 6-bit display according to a time gray scale method as shown in FIG. 45. Each subframe in FIG. 45 is divided into two equally, thereby 12 subframes (SF1 to SF12) are formed and respective lengths of lighting periods thereof are such that SF1=0.5, SF2=1, SF3=2, SF4=4, SF5=8, SF6=16, SF7=0.5, SF8=1, SF9=2, SF10=4, SF11=8, and SF12=16. Note that a gray scale level of 1 and a length of 1 of a lighting period correspond to each other. Similarly to the case of a 5-bit display, gradation is expressed by selecting a subframe to perform lighting at each gray scale level.
As described above, by dividing each subframe into two equally, the frame frequency can be increased to twice substantially.
In addition, as another method for increasing the frame frequency, there has been a method disclosed in Patent Document 3.
Patent Document 3 has described a case of an 8-bit display (256 gray-scale levels). Selection methods of subframes in this case are shown in FIGS. 47A and 47B. In a case of an 8-bit display, according to a conventional time gray scale method, one frame is divided into 8 subframes and respective lengths of lighting periods of the subframes are set so as to be 1, 2, 4, 8, 16, 32, 64, and 128 so that each length of the lighting period is power of two. Described in Patent Document 3 is an example in which only four subframes among the 8 subframes in order of decreasing lighting period are divided; a selection method of subframes in this case is shown in FIG. 47A.
In Patent Document 3, in addition, described is an example in which, in the case of expressing 256 gray-scale levels not by setting each length of the lighting period so as to be power of two but by using an arithmetical progression of which a difference between adjacent bits among 5 higher-order bits is 16 such as that of 1, 2, 4, 8, 16, 32, 48, 64, and 80, only five subframes in order of decreasing lighting period are divided. A selection method of subframes in this case is shown in FIG. 47B.
By using the above-described method, the frame frequency can be increased substantially.
[Patent Document 1] Japanese Patent No. 2903984
[Patent Document 2] Japanese Patent Laid-Open No. 2004-151162
[Patent Document 3] Japanese Patent Laid-Open No. 2001-42818
However, even in the double speed frame method, a pseudo contour occurs where selection of a lighting period is largely changed.
First, a case of a 5-bit display is considered. It is assumed that a gray scale level of 15 is expressed in a pixel A while a gray scale level of 16 is expressed in a pixel B adjacent to the pixel A, using the subframes shown in FIG. 44. A state of lighting/non-lighting in each subframe in that case is shown in FIGS. 48A and 48B. Here, FIG. 48A shows a case of seeing only the pixel A or the pixel B without moving a visual axis. A pseudo contour does not occur in this case. This is because eyes sense brightness in accordance with the sum of brightness where a visual axis passes. Thus, eyes sense that the gray scale level is 15 (=4+2+1+0.5+4+2+1+0.5) in the pixel A and the gray scale level is 16 (=8+8) in the pixel B. That is, an accurate gray scale level is sensed by eyes.
On the other hand, it is assumed that a visual axis moves from the pixel A to the pixel B or from the pixel B to the pixel A. That case is shown in FIG. 48B. In this case, depending on the movement of the visual axis, eyes sense that the gray scale level is 15.5 (=4+2+1+0.5+8) or 23.5 (=8+8+4+2+1+0.5) sometimes. Although it should be seen that the gray scale levels are 15 and 16 normally, the gray scale level is seen to be 15.5 or 23.5 so that a pseudo contour occurs.
Next, a case of a 6-bit display (64 gray-scale levels) is shown in FIG. 49. For example, assuming that a gray scale level of 31 is expressed in a pixel A while a gray scale level of 32 is expressed in a pixel B adjacent to the pixel A, eyes sense that the gray scale level is 31.5 (=8+4+2+1+0.5+16) or 47.5 (=16+16+8+4+2+1+0.5) sometimes, depending on the movement of a visual axis similarly to the case of a 5-bit display. Although it should be seen that the gray scale levels are 31 and 32 normally, the gray scale level is seen to be 31.5 or 47.5 so that a pseudo contour occurs.
Further, the case of FIG. 47A is shown in FIG. 50A and the case of FIG. 47B is shown in FIG. 50B. For example, assuming that a gray scale level of 127 is expressed in a pixel A while a gray scale level of 128 is expressed in a pixel B adjacent to the pixel A, the gray scale level to be sensed is different depending on the movement of a visual axis similarly to the examples described hereinabove. For example, in the case of FIG. 50A, eyes sense that the gray scale level is 121 (=64+32+16+8+1) or 134 (=32+16+8+8+4+2+64) sometimes. In the case of FIG. 50B, eyes sense that the gray scale level is 120 (=40+24+32+16+8) or 134 (=32+16+8+8+4+2+40+24) sometimes. In either case, although it should be seen that the gray scale levels are 127 and 128 normally, the gray scale level is sensed with over width so that a pseudo contour occurs.
In the double speed frame method also, the number of subframes is increased so that a duty ratio (a proportion of a lighting period to one frame) is decreased. Therefore, in order to realize the same average luminance as in the case of not using the double speed frame method, a voltage applied to a light emitting element is increased so that power consumption is increased, reliability of the light emitting element is decreased, and the like.